1. Main points
The 2021 Census of England and Wales included a voluntary question on sexual orientation for the first time; we have linked census data to death registrations and administrative NHS hospital records to estimate how the risk of intentional self-harm and suicide differs by sexual orientation.
The age-standardised rate of intentional self-harm for people who identified with an LGB+ orientation ("Gay or Lesbian", "Bisexual" or "Other sexual orientation") was 1,508.9 per 100,000 people in England and Wales between March 2021 and December 2023, compared with 598.4 per 100,000 people who described themselves as "Straight or Heterosexual".
Compared with people identifying themselves as "Straight or Heterosexual", risk of intentional self-harm was 2.5 times higher for people identifying with an LGB+ orientation, 2.8 times higher for those describing themselves as "Bisexual", 2.4 times higher for those describing themselves as "Gay or Lesbian", and 2.2 times higher for those who selected "Other sexual orientation".
The age-standardised rate of suicide for people identifying with an LGB+ orientation was 50.3 per 100,000 people, compared with 23.1 per 100,000 people describing themselves as "Straight or Heterosexual".
Risk of suicide for people identifying with an LGB+ orientation was 2.2 times higher than for those identifying as "Straight or Heterosexual".
Comparing those identifying with an LGB+ orientation with those identifying as "Straight or Heterosexual", females had a larger relative increase in the risk of both intentional self-harm and suicide than males.
Comparing those identifying with an LGB+ orientation with those identifying as "Straight or Heterosexual", people aged 16 to 24 years had a larger relative increase in the risk of intentional self-harm than all other age groups.
If you are a journalist covering a suicide-related issue, please consider following the Samaritans' media guidelines on the reporting of suicide because of the potentially damaging consequences of irresponsible reporting. In particular, the guidelines advise on terminology and include links to sources of support for anyone affected by the themes in the bulletin.
If you are struggling to cope, please call Samaritans for free on 116 123 (UK and the Republic of Ireland) or contact other sources of support, such as those listed on the NHS help for suicidal thoughts web page. Support is available 24 hours a day, every day of the year, providing a safe place for you, whoever you are and however you are feeling.
For information on the UK government's ambitions and actions towards suicide prevention, see GOV.UK's Suicide prevention strategy for England: 2023 to 2028.
This analysis relates to suicides occurring between 21 March 2021 and 31 December 2023, and registered by 10 February 2025. All deaths by suicide are certified by a coroner, which results in a delay between the date the death occurred and the date of registration. As a result, there may be some suicides occurring in our analysis period that had not been registered by 10 February 2025.
2. Background to the research
The 2021 Census of England and Wales asked people about their sexual orientation for the first time. The sexual orientation question was voluntary and was only asked of people aged 16 years and over. This question did not collect any information on gender identity. In this release, the abbreviation LGB+ refers to people who described their sexual orientation as "Gay or Lesbian", "Bisexual" or "Other sexual orientation" in Census 2021.
We linked Census 2021 data to administrative National Health Service (NHS) hospital records and Office for National Statistics (ONS) death registrations to investigate how rates of NHS hospital admissions and Accident and Emergency (A&E) attendances for intentional self-harm and death by suicide differed by sexual orientation in England and Wales between 21 March 2021 (Census Day) and 31 December 2023. Information on how we defined intentional self-harm and suicide can be found in Section 6: Glossary. Self-harm and suicide were analysed independently.
In this bulletin, we present rates for England and Wales combined. The accompanying dataset includes rates and rate ratios for Wales and regions of England, as well as breakdowns by:
age group
sex
rural or urban classification
ethnic group
country of birth
main language
religion
relative area deprivation
highest level of qualification (for people aged 25 years and over)
National Statistics Socio-economic Classification (NS-SEC)
household size
household tenure
economic activity
self-reported general health
self-reported disability status
residence type
3. Rates of intentional self-harm by sexual orientation
Between 21 March 2021 and 31 December 2023, 135,035 people in our linked study population (0.47% of the study total) were admitted to hospital or attended A&E for intentional self-harm in England and Wales at least once. Of these, 16,680 (12.4%) identified with an LGB+ orientation ("Gay or Lesbian", "Bisexual" or "Other sexual orientation") in Census 2021 and 118,360 (87.7%) identified as "Straight or Heterosexual".
Risk of intentional self-harm was estimated to be 2.5 times greater for people who identified with an LGB+ orientation compared with "Straight or Heterosexual", with an age-standardised rate of 1,508.9 per 100,000 people for the LGB+ population, and an age-standardised rate of 598.4 per 100,000 people for the "Straight or Heterosexual" population. All age-standardised rates throughout this release are standardised to the age distribution of the LGB+ population observed in the linked study dataset.
Within the LGB+ population, risk of intentional self-harm was highest for people describing themselves as "Bisexual" (1,669.5 per 100,000 people; 2.8 times higher than "Straight or Heterosexual"), followed by those describing themselves as "Gay or Lesbian" (1,451.6 per 100,000 people; 2.4 times higher than "Straight or Heterosexual") and those who selected "Other sexual orientation" (1,337.8 per 100,000 people; 2.2 times higher than "Straight or Heterosexual").
We cannot say from this descriptive analysis why rates of intentional self-harm and suicide for people who identify as LGB+ differ to those for people who identify as “Straight or Heterosexual”, and it is not possible to say whether sexual orientation is a causal risk factor for intentional self-harm or suicide.
Rates of intentional self-harm by sexual orientation and sex
Age-standardised rates of intentional self-harm were higher for females (759.4 per 100,000 people) than for males (565.7 per 100,000 people) overall. Both females and males identifying with an LGB+ orientation had a statistically significant higher risk of intentional self-harm than those identifying as "Straight or Heterosexual".
Relative increases in risk of intentional self-harm for the LGB+ population compared with the "Straight or Heterosexual" population were statistically significantly higher for females than males (2.8 times higher for females, and 1.9 times higher for males). Within the LGB+ population, the relative increase in risk for the "Gay or Lesbian" population compared with the "Straight or Heterosexual" population reached 3.0 times higher for females, compared with 1.9 times higher for males. Information on how we defined relative differences in risk in this study is available in the "Rate ratio" definition in Section 6: Glossary.
Figure 1: Comparing those identifying with an LGB+ orientation with those identifying as “Straight or Heterosexual”, females had a statistically significantly larger relative increase in risk of intentional self-harm than males
Age-standardised rates of intentional self-harm per 100,000 people aged 16 years and over by sexual orientation and sex, England and Wales, 21 March 2021 to 31 December 2023
Embed code
Notes
Age-standardised rates of intentional self-harm are expressed per 100,000 people and standardised to the age distribution of the observed study population identifying with an LGB+ orientation in Census 2021.
The error bars are 95% confidence intervals.
All calculations are based on estimates rounded to the nearest multiple of five.
These results are based on those aged 16 years and over who answered the sexual orientation question in Census 2021, completed the census questionnaire themselves and could be linked to an NHS number.
Intentional self-harm relates to NHS hospital admissions and Accident and Emergency (A&E) attendances for intentional self-harm.
Rates of intentional self-harm by sexual orientation and age group
Age-specific rates of intentional self-harm were higher for younger age groups than for older age groups overall, with people aged 16 to 24 years having the highest age-specific rate (1,199.9 per 100,000 people) and people aged 65 years and over having the lowest age-specific rate (212.6 per 100,000 people).
Across all age groups, risk of intentional self-harm was statistically significantly higher for people identifying with an LGB+ orientation than for people identifying as "Straight or Heterosexual".
Within the LGB+ population, those describing themselves as "Bisexual" also had statistically significantly higher risk of intentional self-harm among 25- to 54-year-olds than the "Straight or Heterosexual", "Gay or Lesbian" and "Other sexual orientation" groups.
Relative increases in risk of intentional self-harm for the LGB+ population compared with the "Straight or Heterosexual" group were higher for younger age groups. People aged 16 to 24 years had a statistically significant larger relative increase in risk of intentional self-harm (2.8 times higher for people in the LGB+ population compared with "Straight or Heterosexual") than all other age groups.
Figure 2: Comparing those identifying with an LGB+ orientation with those identifying as “Straight or Heterosexual”, people aged 16 to 24 years had a statistically significantly larger relative increase in the risk of intentional self-harm than all other age groups
Age-specific rates of intentional self-harm per 100,000 people aged 16 years and over by sexual orientation and age group, England and Wales, 21 March 2021 to 31 December 2023
Embed code
Notes
The error bars are 95% confidence intervals.
All calculations are based on estimates rounded to the nearest multiple of five.
These results are based on those aged 16 years and over who answered the sexual orientation question in Census 2021, completed the census questionnaire themselves and could be linked to an NHS number.
Intentional self-harm relates to NHS hospital admissions and Accident and Emergency (A&E) attendances for intentional self-harm.
Rates of self-harm by sexual orientation and other sociodemographic characteristics
Looking across other sociodemographic characteristics, risk of intentional self-harm was consistently statistically significantly higher for the LGB+ population compared with the "Straight or Heterosexual" population.
Groups with the largest relative increases in risk of intentional self-harm for the LGB+ population compared with the "Straight or Heterosexual" population included people:
in the Black ethnic group
with English as their main language
living in rural areas
living in less deprived areas
with a higher National Statistics Socio-economic Classification (NS-SEC)
with a Level 3 highest level of qualification (two or more A Levels or equivalent qualifications)
in good health
not limited by a disability
describing their religion as Buddhist
All the statistical results are available in the accompanying dataset.
Back to table of contents4. Rates of suicide by sexual orientation
Between 21 March 2021 and 31 December 2023, 7,150 people in our linked study population (0.02% of the study total) died by suicide in England and Wales. Of these, 555 (7.8%) identified with an LGB+ orientation ("Gay or Lesbian", "Bisexual" or "Other sexual orientation") in Census 2021 and 6,595 (92.2%) identified as "Straight or Heterosexual".
Risk of suicide was estimated to be 2.2 times greater for people identifying with an LGB+ orientation compared with "Straight or Heterosexual", with an age-standardised rate of 50.3 per 100,000 people for the LGB+ population, and an age-standardised rate of 23.1 per 100,000 people for the "Straight or Heterosexual" population.
Rates of suicide by sexual orientation and sex
Age-standardised rates of suicide were higher for males (37.9 per 100,000 people) than females (13.4 per 100,000 people) overall. Risk of suicide for males was estimated to be 2.8 times higher than females in our linked study population, compared with 3.1 times higher for males than females in the overall suicide rates in England and Wales in 2023.
Both females and males who identified with an LGB+ orientation had a statistically significant higher risk of suicide than those who identified as "Straight or Heterosexual".
Relative increases in risk of suicide for the LGB+ population compared with the "Straight or Heterosexual" population were statistically significantly higher for females than males (3.1 times higher for females, and 1.8 times higher for males).
Figure 3: Comparing those identifying with an LGB+ orientation with those identifying as “Straight or Heterosexual”, females had a statistically significantly larger relative increase in risk of suicide than males
Age-standardised rates of suicide per 100,000 people aged 16 years and over by sex and sexual orientation, England and Wales, 21 March 2021 to 31 December 2023
Embed code
Notes
Age-standardised rates of suicide are expressed per 100,000 people and standardised to the age distribution of the observed study population identifying with an LGB+ orientation in Census 2021.
The error bars are 95% confidence intervals.
All calculations are based on estimates rounded to the nearest multiple of five.
These results are based on those aged 16 years and over who answered the sexual orientation question in Census 2021, completed the census questionnaire themselves and could be linked to an NHS number.
Rates of suicide by sexual orientation and age group
Age-specific rates of suicide were highest for people aged 45 to 54 years (31.1 per 100,000 people) overall. Looking at sexual orientation, people identifying with an LGB+ orientation had statistically significantly higher risk of suicide compared with those identifying as "Straight or Heterosexual" across all age groups.
Relative increases in risk of suicide for the LGB+ population compared with the "Straight or Heterosexual" population were highest for older age groups, with people aged 65 years and over estimated to have 2.8 times higher risk of suicide when identifying with an LGB+ orientation compared with "Straight or Heterosexual".
Figure 4: Across all age groups, people identifying with an LGB+ orientation had a higher risk of suicide compared with those identifying as “Straight or Heterosexual”
Age-specific rates of suicide per 100,000 people aged 16 years and over by sexual orientation and age group, England and Wales, 21 March 2021 to 31 December 2023
Embed code
Notes
The error bars are 95% confidence intervals.
All calculations are based on estimates rounded to the nearest multiple of five.
These results are based on those aged 16 years and over who answered the sexual orientation question in Census 2021, completed the census questionnaire themselves and could be linked to an NHS number.
Rates of suicide by sexual orientation and other sociodemographic characteristics
Looking across other sociodemographic characteristics, risk of suicide was generally numerically higher for the LGB+ population compared with the "Straight or Heterosexual" population. However, we are currently limited in available death registrations data from after 2021 (whereby the cause of death was suicide when broken down by sociodemographic groups) and not all differences were found to be statistically significant.
Groups with the largest relative increases in risk of suicide for the LGB+ population compared with the "Straight or Heterosexual" population included people:
in the Black ethnic group
with English as their main language
with a Level 3 highest level of qualification (two or more A Levels or equivalent qualifications)
in good health
not limited by a disability
All the statistical results are available in the accompanying dataset.
Back to table of contents5. Data on self-harm and suicide by sexual orientation, England and Wales
Self-harm and suicide by sexual orientation, England and Wales
Dataset | Released 9 April 2025
A population-level analysis of rates of intentional self-harm and suicide by sexual orientation for people aged 16 years and over in England and Wales. March 2021 to December 2023.
6. Glossary
Age-specific rate
Age-specific rates are calculated as the total number of people experiencing an outcome (intentional self-harm or suicide) per 100,000 people within an age group, allowing for comparisons of rates between age groups.
Age-standardised rate
Age-standardised rates allow comparisons between populations that contain different proportions of people of different ages. In this analysis, age-standardised rates were calculated as the weighted sum of age-specific rates per 100,000 people in five-year age bands. Rates for the whole study population (those identifying as "Straight or Heterosexual" and those identifying with an LGB+ orientation) were standardised using age-specific weights representing the age distribution of the observed LGB+ study population (which has a younger age distribution than the general population in England and Wales) for comparison. The reported age-standardised rates for the "Straight or Heterosexual" group therefore represent the estimated rate if this group had the same age distribution as that of the LGB+ group.
Confidence interval
A confidence interval is a measure of the uncertainty around a specific estimate. If a confidence interval is calculated at the 95% level, it is expected that the interval will contain the true value on 95 occasions if repeated 100 times. As intervals around estimates widen, the level of uncertainty about where the true value lies increases. The size of the interval around the estimate is strongly related to the number of events of interest (intentional self-harm or suicide), and the size of the underlying population. More information is available on our Uncertainty and how we measure it for our surveys web page.
Rate ratio
Ratios of age-specific and age-standardised rates were calculated as:
Rate ratios represent relative differences in risk of intentional self-harm or death by suicide for the LGB+ group, compared with the "Straight or Heterosexual" group, with ratios bigger than 1 indicating greater risk in the LGB+ group and ratios smaller than 1 indicating greater risk in the "Straight or Heterosexual" group. Rate ratios were also separately calculated for each of the detailed LGB+ groups ("Gay or Lesbian", "Bisexual" and "Other sexual orientation"), each compared with the "Straight or Heterosexual" group. 95% confidence intervals for the rate ratios were calculated through simulation using the standard errors for the age-specific and age-standardised rates.
Self-harm
In this release, self-harm includes all hospital episodes recording:
intentional self-harm (International Classification of Diseases: ICD-10 codes X60 to X84) in the Hospital Episode Statistics (HES) Admitted Patient Care (APC) data from NHS England
intentional self-harm (ICD-10 codes X60 to X84) in the Patient Episode Database for Wales (PEDW) Admitted Patient Care (APC) data from Digital Health and Care Wales
chief complaint code 248062006 ("Self-injurious behaviour (finding)") or injury intent code 276853009 ("Self-inflicted injury (disorder)") in the Emergency Care Data Set (ECDS) from NHS England
attendance group code 13 ("Deliberate self-harm") in the Emergency Department Data Set (EDDS) from Digital Health and Care Wales
Sexual orientation
Sexual orientation is an umbrella term covering sexual identity, attraction and behaviour. For an individual Census 2021 respondent, these may not be the same. For example, someone in an opposite-sex relationship may also experience same-sex attraction, and the other way around. This means the statistics should be interpreted purely as showing how people responded to the census question, rather than necessarily being about whom they are attracted to or their actual relationships.
The sexual orientation groups included in this release were "Straight or Heterosexual", "Gay or Lesbian", "Bisexual" and "Other sexual orientation".
Suicide
In this release, suicide is based on the National Statistics definition of suicide. This includes all deaths from intentional self-harm, and deaths caused by injury or poisoning where the intent was undetermined. Codes corresponding to intentional self-harm (X60 to X84) and injury (including, but not limited to, poisoning) of undetermined intent (Y10 to Y34) from the World Health Organisation's (WHO) International Statistical Classification of Diseases and Related Health Problems, 10th Revision (ICD-10) were used.
Back to table of contents7. Data sources and quality
Data sources
We used anonymised data from Census 2021 as our population base. The Census 2021 data were linked to NHS numbers in the NHS Personal Demographics Service (PDS) 2019 with a linkage rate of 94.6%. More information about Census 2021 linkage to the PDS is available in our Census 2021 to Personal Demographics Service linkage report. To adjust for non-linkage to the NHS PDS, we derived post-stratification weights by sexual orientation, age group, sex, region and ethnic group, which were applied throughout the analysis.
The Census 2021 data were then linked to the following datasets via NHS number:
Office for National Statistics (ONS) death registrations, covering deaths occurring between 21 March 2021 and 31 December 2023 and registered by 10 February 2025
Hospital Episode Statistics (HES) Admitted Patient Care (APC) from NHS England
Emergency Care Data Set (ECDS) from NHS England
Patient Episode Database for Wales (PEDW) Admitted Patient Care (APC) from Digital Health and Care Wales
Emergency Department Data Set (EDDS) from Digital Health and Care Wales
Around 76.3% of suicides in the ONS death registrations and 79.6% of intentional self-harm episodes in the NHS hospital records could be linked to the NHS PDS.
Inclusion criteria
Individuals were included in our study population if they had a record in the Census 2021 dataset and had a valid NHS number linked to their Census ID. The population was then restricted to individuals who:
were aged 16 years and over on Census Day (21 March 2021)
were either alive at the end of the study (31 December 2023) or died between 21 March 2021 and the end of the study
answered the sexual orientation question in Census 2021
completed the census questionnaire themselves (not by proxy)
Around 92.5% of Census 2021 respondents aged 16 years and over answered the question on sexual orientation, while the remaining 7.5% chose not to. Approximately 30% of responses to Census 2021 were given by proxy.
Our final study population comprised 28,659,465 individuals (59.0% of all people aged 16 years and over in Census 2021), who answered the sexual orientation question in Census 2021, completed the census questionnaire themselves, and could be linked to the NHS's 2019 Personal Demographics Service (PDS). Around 135,035 individuals were identified as having at least one hospital admission or A&E attendance for intentional self-harm during the study period, and 7,150 individuals were identified as dying by suicide during the study period. More information can be found in the sample flow in our accompanying dataset.
In our linked study population, 27,554,145 people (96.1%) identified as "Straight or Heterosexual" and 1,105,325 people (3.9%) identified with an LGB+ orientation. Within the LGB+ population, 524,585 people described themselves as "Gay or Lesbian", 457,520 people described themselves as "Bisexual", and 123,220 people selected "Other sexual orientation".
For comparison, of the 92.5% of people who chose to answer the sexual orientation question in the full Census 2021 dataset, 96.6% of people identified as "Straight or Heterosexual" and 3.4% of people identified with an LGB+ orientation.
All sociodemographic variables included in the analyses were self-reported in Census 2021.
Follow-up
Individuals were followed up from 21 March 2021 (Census Day) to 31 December 2023, unless they died during this period (in which case, they were followed up to date of death).
Strengths and limitations
The main strength of this analysis is the use of a population-level linked dataset, covering all usual residents of England and Wales aged 16 years and over who answered the sexual orientation question in Census 2021, completed the census questionnaire themselves, and could be linked to an NHS number. This is the largest study to examine rates of self-harm and suicide by sexual orientation to date.
The census provides high quality information about characteristics such as sexual orientation, socioeconomic classification, self-reported ethnicity and general health. Census 2021 covered an estimated 97% of the population, and therefore is the most representative data source available to produce statistics about the population living in England and Wales.
However, not all people living in England and Wales in March 2021 were enumerated at Census 2021 (for example, because of non-response), and of those who were, not all could be linked to an NHS number via the Personal Demographics Service (PDS). To mitigate against linkage bias, we derived post-stratification weights by sexual orientation, age group, sex, region and ethnic group, which were applied throughout the analysis. Quality considerations along with the strengths and limitations of Census 2021 more generally can be found in our Quality and Methodology Information report (QMI) for Census 2021.
We excluded proxy Census 2021 responses from our study population to mitigate against possible respondent bias in reporting of sexual orientation. However, this may increase bias in other respects, because there are higher proportions of proxy responses in some sociodemographic groups, for example, those in bad health and those with lower-level qualifications. To check robustness of our findings, we included proxy responses in a sensitivity analysis and found that trends remained similar.
Our estimates of self-harm reflect the number of people who were admitted to an NHS hospital or attended A&E for intentional self-harm in England or Wales. It is important to note that this does not reflect the true prevalence of intentional self-harm, as not all people who intentionally self-harm will attend hospital, some people may attend hospital for intentional self-harm under private healthcare, and some NHS hospital records for self-harm may not have recorded intent. Determining intent from administrative health records can be challenging, with 17.9% of Emergency Care Data Set (ECDS) records in England missing data on both chief complaint and injury intent, and 36.4% of Emergency Department Data Set (EDDS) records in Wales missing data on reason for admission or recording an "unknown" reason for admission.
We looked at hospital admissions and A&E attendances for intentional self-harm as a binary measure, which does not discriminate between people who attended hospital for intentional self-harm only once, and those who attended hospital multiple times, or the length of stay in hospital once admitted. Therefore, our analysis may not reflect potential compounding effects of sexual orientation on the likelihood of attending hospital for intentional self-harm multiple times, or the overall healthcare utilisation associated with intentional self-harm.
Finally, we cannot say from this descriptive analysis why rates of intentional self-harm and suicide for people who identify as LGB+ differ to those for people who identify as "Straight or Heterosexual", and it is not possible to say whether sexual orientation is a causal risk factor for intentional self-harm or suicide.
Collaboration
This analysis was produced in collaboration with King's College London, with input from the Office for Health Improvement and Disparities (OHID).
Back to table of contents9. Cite this statistical bulletin
Office for National Statistics (ONS), released 9 April 2025, ONS website, statistical bulletin, Self-harm and suicide by sexual orientation, England and Wales: March 2021 to December 2023